我国农药过量使用造成严重环境问题，减施增效的变量作业是现代智能农机发展方向，基于无人机低空遥感技术进行病害信息快速定位与检测有较好应用前景。本研究以水稻典型病害为研究对象，采用无人机搭载高光谱、多光谱、RGB相机与热红外相机进行田间遥感数据获取，对数据进行太阳辐照校正、大气修正等预处理之后，结合经RTK GNSS/INS深耦合计算所得外方位元素进行直接地理配准，再依据地面验证数据建立特征级多源数据融合的水稻病害检测模型，研究多源遥感数据多尺度分解的图像融合方法。本研究将丰富水稻病害检测手段，提高大田病害检测效率，研究所得结果将为变量施药提供数据支撑。

The excessive use of pesticides in China has caused serious environmental problems. Variable operation of reducing the application and increasing efficiency is the development direction of modern intelligent agricultural machinery. Therefore, rapid positioning and detection of disease information based on UAV low-altitude remote sensing technology has a good application prospect. In this study, the typical diseases of rice were studied. The UAV was equipped with hyperspectral, multi-spectral, RGB camera and thermal infrared camera to acquire the remote sensing data in the field. After the data was preprocessed through solar irradiation correction and atmospheric correction etc., the data was georeferenced by the result of RTK GNSS/INS deep coupling calculation of the external orientation elements for direct georeferencing. And then based on ground verification data to establish, a multi-source data fusion method is built for multi-source remote sensing data fusion model. This study will enrich the detection methods of rice diseases and improve the detection efficiency of field diseases. The results of the study will provide data support for variable pesticide application.